Quantitative measurement of tumor biomarkers has shown great promise in predicting patient outcome and response to treatment. Recently, we have developed an automated system for assessing biomarker expression on tissue sections (AQUA - Automated Quantitative Analysis). AQUA provides quantitative analysis and sub-cellular localization of biomarkers on immunohistochemically stained tissues. AQUA, however, is limited by 1) the resolution of light microscopy and 2) the non-linear effects of enzymatic amplification used to identify biomarkers. The progression toward bio-specific therapies and associated pharmaco-diagnostics will require linear, quantitative measures of protein expression within each patient's tumor. Individual tumor profiling can also be advanced by developing methods for assessing the functioning of tumor-related pathways, in particular by studying specific protein-protein interactions. To acomplish this, we propose the following specific aims: 1: To develop methods for identifying specific protein-protein interactions in tissue sections. And 2: To develop methods for increasing the dynamic range of assessing biomarker expression. To assess protein-protein interactions - which are far below the resolution of light microscopy - we will develop new techniques that combine heterobifunctional crosslinking reagents, fluorescence energy resonance transfer (FRET), and catalyzed tyramide-amplification, making them suitable for use in tissue sections. To improve the dynamic range of the AQUA technology, we will adapt several recently described methods for building large fluorescent molecules in situ (Christmas trees, rolling-circle and ImmunoAT-tailing). Once these techniques are developed, they will provide new linear, quantitative, and highly specific assessments of biomarkers and pathways in tissues that can ultimately be used to advance personalized pharmaco-diagnostics. [unreadable] [unreadable] [unreadable]